A closer look at NFV execution models

Peng Zheng; Arvind Narayanan; Zhi Li Zhang

doi:10.1145/3343180.3343188

A closer look at NFV execution models

Peng Zheng, Arvind Narayanan, Zhi Li Zhang

Computer Science and Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

11 Scopus citations

Abstract

Network Function Virtualization (NFV) advocates running service function chains (SFCs) on commodity servers as software, thereby providing a new level of flexibility to the deployment and management of network services. However, as we move from 10/40 Gbps to 100/400 Gbps line rates, it is challenging to build an NF execution framework that can deliver high performance at the maximum line speed using commodity servers, while providing scalability and flexibility afforded by software. In this paper, we investigate a fundamental problem of any NFV framework, i.e. how to execute SFCs on commodity servers by examining and comparing the performance of two execution models: the pipeline and run-to-completion models. In particular, we investigate how the multi-core server architecture affects the performance of SFC execution models by conducting extensive experiments on a testbed and shed new insights on the design and optimization of SFC execution models.

Original language	English (US)
Title of host publication	APNet 2019 - Proceedings of the 2019 Asia-Pacific Workshop on Networking
Publisher	Association for Computing Machinery
Pages	85-91
Number of pages	7
ISBN (Electronic)	9781450376358
DOIs	https://doi.org/10.1145/3343180.3343188
State	Published - Aug 17 2019
Event	3rd ACM SIGCOMM Asia-Pacific Workshop on Networking, APNet 2019 - Beijing, China Duration: Aug 17 2019 → Aug 18 2019

Publication series

Name	ACM International Conference Proceeding Series

Conference

Conference	3rd ACM SIGCOMM Asia-Pacific Workshop on Networking, APNet 2019
Country/Territory	China
City	Beijing
Period	8/17/19 → 8/18/19

Bibliographical note

Funding Information:
The research is supported in part by US NSF under Grants CNS-1411636, CNS-1618339, CNS-1617729, CNS-1814322, and CNS-1836772. Peng Zheng gratefully acknowledges financial support from National Key R&D Program of China (2017YFB0801703) and China Scholarship Council.

Publisher Copyright:
© 2019 Association for Computing Machinery.

Keywords

Execution model
Hybrid
Multi-core
NFV
NUMA
Pipeline
Run-to-completion

Publisher link

10.1145/3343180.3343188

Find It

Find It at U of M Libraries

Cite this

Zheng, P, Narayanan, A & Zhang, ZL 2019, A closer look at NFV execution models. in APNet 2019 - Proceedings of the 2019 Asia-Pacific Workshop on Networking. ACM International Conference Proceeding Series, Association for Computing Machinery, pp. 85-91, 3rd ACM SIGCOMM Asia-Pacific Workshop on Networking, APNet 2019, Beijing, China, 8/17/19. https://doi.org/10.1145/3343180.3343188

@inproceedings{2d8ac55dad544f5eaca7de19b3022fdd,

title = "A closer look at NFV execution models",

abstract = "Network Function Virtualization (NFV) advocates running service function chains (SFCs) on commodity servers as software, thereby providing a new level of flexibility to the deployment and management of network services. However, as we move from 10/40 Gbps to 100/400 Gbps line rates, it is challenging to build an NF execution framework that can deliver high performance at the maximum line speed using commodity servers, while providing scalability and flexibility afforded by software. In this paper, we investigate a fundamental problem of any NFV framework, i.e. how to execute SFCs on commodity servers by examining and comparing the performance of two execution models: the pipeline and run-to-completion models. In particular, we investigate how the multi-core server architecture affects the performance of SFC execution models by conducting extensive experiments on a testbed and shed new insights on the design and optimization of SFC execution models.",

keywords = "Execution model, Hybrid, Multi-core, NFV, NUMA, Pipeline, Run-to-completion",

author = "Peng Zheng and Arvind Narayanan and Zhang, {Zhi Li}",

note = "Funding Information: The research is supported in part by US NSF under Grants CNS-1411636, CNS-1618339, CNS-1617729, CNS-1814322, and CNS-1836772. Peng Zheng gratefully acknowledges financial support from National Key R&D Program of China (2017YFB0801703) and China Scholarship Council. Publisher Copyright: {\textcopyright} 2019 Association for Computing Machinery.; 3rd ACM SIGCOMM Asia-Pacific Workshop on Networking, APNet 2019 ; Conference date: 17-08-2019 Through 18-08-2019",

year = "2019",

month = aug,

day = "17",

doi = "10.1145/3343180.3343188",

language = "English (US)",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

pages = "85--91",

booktitle = "APNet 2019 - Proceedings of the 2019 Asia-Pacific Workshop on Networking",

}

TY - GEN

T1 - A closer look at NFV execution models

AU - Zheng, Peng

AU - Narayanan, Arvind

AU - Zhang, Zhi Li

N1 - Funding Information: The research is supported in part by US NSF under Grants CNS-1411636, CNS-1618339, CNS-1617729, CNS-1814322, and CNS-1836772. Peng Zheng gratefully acknowledges financial support from National Key R&D Program of China (2017YFB0801703) and China Scholarship Council. Publisher Copyright: © 2019 Association for Computing Machinery.

PY - 2019/8/17

Y1 - 2019/8/17

N2 - Network Function Virtualization (NFV) advocates running service function chains (SFCs) on commodity servers as software, thereby providing a new level of flexibility to the deployment and management of network services. However, as we move from 10/40 Gbps to 100/400 Gbps line rates, it is challenging to build an NF execution framework that can deliver high performance at the maximum line speed using commodity servers, while providing scalability and flexibility afforded by software. In this paper, we investigate a fundamental problem of any NFV framework, i.e. how to execute SFCs on commodity servers by examining and comparing the performance of two execution models: the pipeline and run-to-completion models. In particular, we investigate how the multi-core server architecture affects the performance of SFC execution models by conducting extensive experiments on a testbed and shed new insights on the design and optimization of SFC execution models.

AB - Network Function Virtualization (NFV) advocates running service function chains (SFCs) on commodity servers as software, thereby providing a new level of flexibility to the deployment and management of network services. However, as we move from 10/40 Gbps to 100/400 Gbps line rates, it is challenging to build an NF execution framework that can deliver high performance at the maximum line speed using commodity servers, while providing scalability and flexibility afforded by software. In this paper, we investigate a fundamental problem of any NFV framework, i.e. how to execute SFCs on commodity servers by examining and comparing the performance of two execution models: the pipeline and run-to-completion models. In particular, we investigate how the multi-core server architecture affects the performance of SFC execution models by conducting extensive experiments on a testbed and shed new insights on the design and optimization of SFC execution models.

KW - Execution model

KW - Hybrid

KW - Multi-core

KW - NFV

KW - NUMA

KW - Pipeline

KW - Run-to-completion

UR - http://www.scopus.com/inward/record.url?scp=85072037820&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072037820&partnerID=8YFLogxK

U2 - 10.1145/3343180.3343188

DO - 10.1145/3343180.3343188

M3 - Conference contribution

AN - SCOPUS:85072037820

T3 - ACM International Conference Proceeding Series

SP - 85

EP - 91

BT - APNet 2019 - Proceedings of the 2019 Asia-Pacific Workshop on Networking

PB - Association for Computing Machinery

T2 - 3rd ACM SIGCOMM Asia-Pacific Workshop on Networking, APNet 2019

Y2 - 17 August 2019 through 18 August 2019

ER -

A closer look at NFV execution models

Abstract

Publication series

Conference

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this